On 05/26/2014 10:49 AM, Viresh Kumar wrote:
On 25 May 2014 15:08, Preeti U Murthy preeti@linux.vnet.ibm.com wrote:
So I would suggest the below for this problem:
Add an additional case CLOCK_EVT_MODE_ONESHOT_STOPPED, which returns an error code. Let the 'default' case stay as it is and return success just like it is implemented today.
Its not a good idea to bug all drivers for every new mode that comes in. Let drivers only take care of modes they are concerned with and for rest they should return -ENOSYS.
Returning success from 'default' look like a poor design to me.
Ah ok! I see now. I mistook the changelog. I assumed that these drivers are returning success in default case although they are disabling the clock device and that we now intend to change that to return an error code. Sorry about this!
Now I see that it is an issue. Now we are going to hit the default case and get returned an error code, but with disabled clock device which is unexpected.Hmmm..this needs some thought.
Regards Preeti U Murthy